Bacteria can rapidly evolve mechanisms allowing them to use toxic environmental pollutants as a carbon source. In the current study we examined whether the survival and evolution of indigenous bacteria with the capacity to degrade organic pollutants could be connected with increased mutation frequency. The presence of constitutive and transient mutators was monitored among 53 pollutants-degrading indigenous bacterial strains. Only two strains expressed a moderate mutator phenotype and six were hypomutators, which implies that constitutively increased mutability has not been prevalent in the evolution of pollutants degrading bacteria. At the same time, a large proportion of the studied indigenous strains exhibited UV-irradiation-induced mutagenesis, indicating that these strains possess error-prone DNA polymerases which could elevate mutation frequency transiently under the conditions of DNA damage. A closer inspection of two Pseudomonas fluorescens strains PC20 and PC24 revealed that they harbour genes for ImuC (DnaE2) and more than one copy of genes for Pol V. Our results also revealed that availability of other nutrients in addition to aromatic pollutants in the growth environment of bacteria affects mutagenic effects of aromatic compounds. These results also implied that mutagenicity might be affected by a factor of how long bacteria have evolved to use a particular pollutant as a carbon source.
Citation: Ilmjärv T, Naanuri E, Kivisaar M (2017) Contribution of increased mutagenesis to the evolution of pollutants-degrading indigenous bacteria. PLoS ONE 12(8): e0182484. https://doi.org/10.1371/journal.pone.0182484
Editor: Digby F. Warner, University of Cape Town, SOUTH AFRICA
Received: October 31, 2016; Accepted: July 19, 2017; Published: August 4, 2017
Copyright: © 2017 Ilmjärv et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All relevant data are within the paper and its Supporting Information files. The nucleotide sequences analysed in the current study are deposited in GenBank (see S7 Table for GenBank accession numbers).
Funding: This work was supported by Institutional Research Funding IUT20-19 from Eesti Teadusagentuur (Estonian Research Council) to MK. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Contamination of the environment with various organic xenobiotic compounds is a widespread problem throughout the world, particularly in industrialized areas. Among the most abundant environmental pollutants, aromatic compounds are of major concern because of their persistence and toxicity. Although many of these compounds are resistant to degradation, the large genetic plasticity and metabolic versatility of bacteria allows them to evolve the necessary catabolic activities to use aromatic compounds as the sole carbon and energy source. New catabolic pathways evolve rapidly as a result of horizontal transfer of genes and point mutations that broaden the substrate range of pre-existing enzymes [1–6].
Most mutations are likely to be deleterious, and so the spontaneous mutation rate is generally held at a low level . However, bacterial populations with higher mutation rates can adapt to novel environments faster than those with lower mutation rates . The mutation rate can be elevated either constitutively (constitutive mutators) or transiently, in response to DNA damage or other stress situations [9, 10]. Transiently increased mutability could be the consequence of the action of the specific low-fidelity DNA polymerases, e.g., Pol IV, Pol V and DnaE2 [11, 12]. Depending on the type of function that is defective, constitutive mutators can have mutation rates that are moderately (about 10-fold) to strongly (100-1000-fold) increased . The most potent constitutive mutators are bacteria defective in either the DNA polymerase III proofreading subunit or the methyl-directed mismatch repair system . Mutators are expected to be rare in well-adapted populations . However, surveys of natural isolates of pathogenic and commensal bacteria have revealed that constitutive mutators exist in the environment at much higher frequencies [14–16]. Subsequent studies [reviewed in [17, 18]] demonstrated that hypermutators can be readily isolated from populations of a variety of human pathogens at frequencies ranging from 1% to 58% of total population. The high proportion of mutators among clinical isolates can be explained by advantage that they give in generating the variability pool for selecting for mutations that could be beneficial under harsh environmental conditions bacteria meet during pathogenesis. By selecting a favourable allele, natural selection indirectly selects the DNA in which this mutation occurred. Consequently, mutator alleles can spread in microbial populations by hitchhiking: mutators can acquire favourable alleles more frequently than nonmutators and will therefore increase in frequency if the advantage of beneficial alleles is greater than the cost of being a mutator . So far, surveys of constitutive mutators have mostly included human pathogens. Whether the survival and evolution of organic pollutants degrading indigenous bacteria could also be connected with mutator phenotype is still unexplored.
Above a certain threshold organic compounds are toxic to bacteria as they damage cell membrane and impair biosynthetic pathways [20, 21]. Membrane stress is connected with generation of reactive oxygen species (ROS) [22, 23]. Notably, exposure of Pseudomonas putida to toluene induces transcription of uvrA and uvrB genes that are involved in nucleotide excision repair (NER) , which indicates that the exposure of bacteria to pollutants could induce DNA damage. Thus, exposure of bacteria to aromatic pollutants and aliphatic solvents may facilitate generation of new catabolic pathways due to their potential mutagenic effects. This idea has been strongly supported by the recent finding that the faulty dioxygenation reaction of the evolving enzyme generates ROS, which, in turn, elevates mutation frequency in the 2,4-dinitrotoluene-degrading bacteria .
The aim of the current study was to explore mechanisms of evolution of xenobiotics degrading indigenous bacteria. Specifically, we assayed for the presence of constitutive and transient mutators among pollutants degrading bacteria, many of them isolated from heavily contaminated environments.
Estimation of spontaneous mutation frequency among pollutants-degrading indigenous strains disclosed rare appearance of constitutive mutators
In order to examine the possibility that the evolution of xenobiotic degrading bacteria could be connected with the appearance of constitutive mutators, we estimated the spontaneous mutation frequencies of 53 indigenous strains, mostly pseudomonads, which were able to degrade various pollutants, e.g., alkanes and phenolic compounds (Table 1). These strains were originated from the Baltic Sea and from the river water polluted with phenolic compounds of the oil shale industry leachate in Northeast Estonia [24–26]. In order to search for constitutive mutators, we employed the rpoB/Rifr test system which has been commonly used to estimate the frequency of mutations in Gammaproteobacteria, e.g., in Escherichia coli  and in pseudomonads . Specifically, we performed fluctuation tests and determined the frequency of rifampicin-resistant mutant colonies/ml with regard to the total viable colony count/ml of the same culture. Spontaneous Rifr mutant frequency of the indigenous strains was compared with that of the laboratory reference strain P. putida PaW85. P. putida strain PaW85  is a plasmid-free derivative of the toluene catabolic plasmid pWW0-harbouring strain mt-2 (PaW1) . This strain is isogenic to the strain KT2440 which is widely used in laboratory experiments . According to our previous studies  PaW85 has a modal or so called “normal” Rifr mutant frequency. The median value of the Rifr mutant frequency (f) of PaW85 estimated in the current study was 1.6 x 10−8. Approximately 87% of the studied indigenous isolates had the Rifr mutant frequency statistically similar to that of the reference strain PaW85. Only two of the studied strains were moderate mutators with considerably higher mutant frequency in comparison with PaW85 (Fig 1 and S1 Table). Specifically, P. stutzeri strain 2C23 and P. putida strain 2D61 exhibited about 15- and 6-fold higher Rifr frequency than the reference strain P. putida PaW85 (p = 0.0094 and p<0.0001, respectively). However, according to the Benjamini-Hochberg procedure only the strain 2D61 had statistically higher mutant frequency. Notably, out of all studied strains about 11% (6 strains) were hypomutators. The hypomutators were represented among most of the studied species except P. stutzeri.
Presented are only the results for strains exhibiting the mutant frequency statistically significantly different from that of the reference strain PaW85 (the Kruskal-Wallis test, p<0.006; S1 Table). Isolate D2RT is missing from the figure because the median value of the frequency of Rifr mutants of this strain was 0 (the mean value of Rifr frequency of this strain was 1.07 x 10−9).
To evaluate the results obtained with the rpoB/Rifr assay, we repeated the estimation of the mutant frequency by using another assay which monitored the appearance of spontaneous streptomycin resistant mutants (Smr assay; S1 Fig). In addition, we excluded a possibility that the estimated mutant frequencies of the studied environmental isolates could be affected by variations in the tolerance to Rif or Sm by measuring the minimal inhibitory concentrations (MIC) (S2 Table). Representatives from different species either exhibiting the highest or the lowest Rifr mutant frequency in comparison with the other strains from the same species were included in this assay. In general, the strains with the highest Rifr mutant frequency exhibited also an elevated Smr mutant frequency (S1 Fig). This assay also disclosed the absence of strong mutators among the studied indigenous strains. Notably, although the Rifr and Smr mutant frequencies of the reference strain PaW85 were comparable, the majority of the studied indigenous strains displayed statistically significantly lower frequency of Smr mutants in comparison to the frequency of Rifr mutants (S1 Fig, S3 and S4 Tables). Taken together, the absence of strong mutators among the studied indigenous strains implied that constitutive mutability could have only a minor role on the evolution of pollutants-degrading indigenous strains.
Indigenous isolates harboured DNA damage-induced mutagenic pathways
In addition to the constitutive pathway increased mutability can be transient and inducible, for example, through the action of specific low-fidelity DNA polymerases [11, 18]. One of the best studied is the UV irradiation-inducible DNA polymerase Pol V-dependent mutagenic pathway encoded by the umuDC operon in E. coli [32–34]. In many environmental bacteria, e.g., in P. putida and P. syringae the Pol V genes (usually named as rulAB genes) are frequently located on plasmids . The translesion DNA synthesis (TLS) provided by the mutagenic DNA polymerase Pol V grants bacteria higher tolerance to UV-C irradiation and to several other DNA-damaging agents [32, 34]. For instance, we have previously shown that the insertion of the rulAB mutagenic operon from the toluene catabolic plasmid pWW0 into the chromosome of P. putida PaW85 (strain PaWrulAB) increased approximately 10 times the tolerance against the UV-C irradiation at a dose of 50 J/m2 . The UV tolerance and mutability phenotype is widely distributed among various bacterial species [18, 36]. Usually there are no quantitative data available how frequent this phenotype is. However, it was reported that about 50% of Pseudomonas syringae pathovars isolated from plants throughout the world expressed the UV tolerance phenotype and carried the rulAB genes . Therefore, we expected that some of the indigenous bacterial strains characterized in this study could also harbour DNA damage-inducible mutagenesis pathways. Since the higher tolerance to UV-C irradiation could hint to the presence of DNA damage-induced mutagenic pathways, we estimated the UV tolerance of the isolated strains. The survival rate of bacteria was measured after the exposure of cells to the UV-C light at doses of 10 J/m2 and 20 J/m2. Two P. putida strains, PaWrulAB and PaW1, both of which possessed the Pol V-encoding rulAB operon, were used as positive controls in the UV tolerance experiments.
We observed that ~89% of the studied strains (41 strains from the total of 46 investigated) had statistically similar tolerance to the UV irradiation (20 J/m2) in comparison with the strain PaW1 (Table 2 and S5 Table). In addition to P. stutzeri we were able to detect bacterial isolates with increased UV tolerance among the species of P. fluorescens and Acinetobacter johnsonii. All P. putida indigenous isolates exhibited lower UV tolerance than the P. putida laboratory strains PaWrulAB and PaW1 (S5 Table). As already mentioned above, the strains with increased tolerance against the UV irradiation might be potential candidates for possessing DNA damage-induced mutagenic pathways. In order to evaluate this possibility, we measured the UV mutagenesis in the strains which had at least 10-fold higher tolerance to the UV irradiation in comparison with the reference strain PaW85. Bacteria were treated by UV-C irradiation at a dose of 100 J/m2 which has been previously used in the UV mutagenesis assay . P. putida strains PaWrulAB and PaW1 were used as positive controls. Under such conditions the exposure of the reference strain PaW85 to the UV irradiation had no increasing effect on the frequency of Rifr mutants (Fig 2, p>0.05). At the same time, the presence of the rulAB operon increased the Rifr mutant frequency more than 10 times, which was even higher than that observed in our previous experiments . About 90% of the studied indigenous isolates with the increased UV tolerance expressed also the UV mutagenesis phenotype (Fig 2 and S6 Table). In addition, we estimated the effect of the UV irradiation on the Rifr mutant frequency in the case of 6 strains (strains PC16, PC18, PC20, PC24, D14, and 2D61) which exhibited lower tolerance against the UV irradiation. It turned out that 3 strains (PC16, PC18, and PC24) exhibited the UV mutagenesis phenotype (S6 Table). Thus, our results indicated that at least one third of the indigenous strains examined in this study have a potential to evolve via transient mutability. However, the real proportion of such strains could be much larger since based on our results the increased UV tolerance seems to be not the only indicator of the presence of mutagenic pathways in the studied isolates.
Presented are only the results with indigenous strains exhibiting statistically significantly increased mutation frequency (p<0.045) after the exposure to UV-C irradiation analysed with Mann-Whitney U test (S6 Table).
The genomes of P. fluorescens strains PC20 and PC24 possess several copies of Pol V genes and a copy of imuC
Among the strains investigated in this study, P. fluorescens strains PC20 and PC24 have been shown to have a great potential for application in bioaugmentation and rhizoremediation [38, 39]. Therefore, to further explore the catabolic potential of PC20 and PC24, the genomes of these strains have been sequenced in Ain Heinaru’s laboratory. The availability of the draft sequence data enabled us a closer inspection for the presence of error-prone DNA polymerases in these strains. Importantly, in a number of bacterial species which do not have the umuDC genes or their orthologs (e.g., rulAB) in the chromosome, carry instead another DNA damage-regulated operon, originally named as imuABdnaE2 operon (renamed later as imuABC operon), which encodes for a mutagenic paralog of the replicative DNA polymerase Pol III, named as DnaE2 or ImuC . This DNA polymerase has been shown to be responsible for the UV tolerance and UV-induced mutagenesis in various bacterial species, e.g., in Mycobacterium tuberculosis and in Caulobacter crescentus [40–42]. The DNA damage-inducible imuABC operon has been identified also in pseudomonads . Therefore, we decided to analyse the sequence data of the strains PC20 and PC24 not only for the presence of genes encoding for Pol V but also for the presence of ImuC. We found that both strains harboured the imuABC operon. Surprisingly, the strain PC20 possessed additionally three copies of the rulAB operon, two in the chromosome and the third one was localized on the plasmid pG20 (unpublished data, GenBank accession number KX893538). The strain PC24 carried two copies of the rulAB operon in the chromosome.
Further characterization of the putative Pol V genes from PC20 and PC24 revealed that the rulAB genes, which are located on the plasmid pG20 of the strain PC20, were clustered together with the rulAB sequences present on the catabolic plasmids pWW0 and NAH7 (Fig 3 and S7 Table). The two chromosomal copies, the rulAB1 and rulAB2 genes of the strain PC20 also clustered together, but were distantly related to the rulABpG20. Meanwhile, the constructed phylogenetic tree revealed that the two copies of the rulAB genes identified in the chromosome of the strain PC24 were less related to each other. The rulAB1operon of PC24 was more closely clustered to the umuDC operon of E. coli K-12 and the rulABpG20, but the rulAB2 operon was more related to the rulAB copies identified in the strain PC20.
The phylogenetic trees of RulA (A) and RulB (B) protein sequences. Underlining indicates the genes of proteins which are located on plasmids. Sequences derived from P. fluorescens strains PC20 and PC24 are shown in boldface. Sequences were aligned with ClustalX2 and further visualised with TreeViewX. Numbers at branch nodes indicate bootstrapping values for 1000 bootstrap replicates. Values under 500 were removed. The identifiers of the aligned sequences are presented in S7 Table.
Multiple sequence alignments of the RulA and RulB homologs from the strains PC20 and PC24 with the sequence of the biochemically and structurally well characterized DNA polymerase Pol V from E. coli revealed that the sequences of the RulA and RulB from PC20 and PC24 contained all Pol V domains. Moreover, all of the analysed sequences with the only exception of the RulB sequence at the amino acid position 72 contained the same residues which have been shown to be essential for the catalytic activity of the umuDC-encoded Pol V in E. coli (S2 and S3 Figs, S7 Table).
Construction of the phylogenetic tree of the deduced amino acid sequences of the imuC from PC20 and PC24 and from other pseudomonads demonstrated that the ImuC homologs from PC20 and PC24 were more closely clustered with the homologs from P. fluorescens than with those from the other Pseudomonas species (Fig 4 and S7 Table). Based on multiple sequence alignment and according to the NCBI Conserved Protein Family database the ImuC homologs of PC20 and PC24 carry functional domains and putative active site residues of ImuC/DnaE2 polymerases (S4 Fig and S7 Table). Thus, the results of DNA sequence analysis indicated that most likely both PC20 and PC24 contain more than one copy of functional Pol V genes and a copy of imuC.
The underlined proteins are from strains possessing rulAB genes and the ones marked with star are from strains where rulAB genes are located in the chromosome. Sequences derived from P. fluorescens strains PC20 and PC24 are shown in boldface. Sequences were aligned with ClustalX2 and further visualised with TreeViewX. Numbers at branch nodes indicate bootstrapping values for 1000 bootstrap replicates. Values under 500 were removed. The identifiers of the aligned sequences are presented in S7 Table.
The rulB and imuC homologs were induced under the conditions of DNA damage
Although the strains PC20 and PC24 carried more than one copy of the rulAB genes in addition to the imuC gene, they were more sensitive to the UV irradiation than the strains PaW1 and PaWrulAB (Table 2). To further examine functionality of these putative DNA polymerase genes, we investigated whether all copies of the rulAB genes and the imuC gene were expressed under the conditions of exogenously induced DNA damage. We have previously shown that the transcription of the rulAB genes located on the plasmid pWW0 can be induced by the DNA damage-inducing chemical mitomycin C (MMC) . Transcription from the LexA2-regulated promoter of the “mutagenesis cassette” encoding for ImuC polymerase in P. putida was also induced in the presence of MMC . According to our previous studies the treatment of cells with MMC gives more reproducible results on gene expression profile than the exposure of bacteria to UV irradiation . Therefore, in order to evaluate the expression of the putative DNA polymerase genes in the strains PC20 and PC24, a quantitative reverse transcriptase polymerase chain reaction (RT-qPCR) analysis was performed using RNA extracted from cells grown in the presence or absence of MMC (experimental details are given in Materials and Methods section). As a positive control we employed the P. putida strain PaW1 which carried the MMC-inducible rulAB operon on the TOL plasmid pWW0 and the imuC gene in the chromosome. The results shown in Fig 5 revealed that under such conditions the transcription of the rulABpWW0 and the imuC gene were induced in the positive control strain PaW1 about 14 fold and 8 fold, respectively. The transcription of all copies of the rulAB operons as well the imuC gene in the strains PC20 and PC24 were also induced by MMC (Fig 5 and S8 Table). However, the induction levels of these genes were lower in comparison with the control strain (Fig 5A). At the same time, the basal level of transcription of the rulAB genes in the strains PC20 and PC24 was higher than that estimated in the reference strain PaW1 (Fig 5B).
(A) The results are presented as relative fold difference between cells with and without MMC treatment. (B) The expression rate of the studied error-prone DNA polymerases normalized against trpA. Each data represents the mean from at least three replicate induction experiments.
These results altogether suggested that the genes encoding for putative error-prone DNA polymerases Pol V and ImuC in the strains PC20 and PC24 could be functional and inducible under the conditions of DNA damage. Consequently, it was hard to understand why the UV mutagenesis phenotype, although quite modest, appeared only in the case of the strain PC24 (S6 Table). Meanwhile, examination of UV mutagenesis phenotype of the strain PC20 was complicated because the survival of this strain upon exposure to the UV-C irradiation at a dose of 100 J/m2 was very low. Thus, it is possible that the experimental conditions which were set up based on the commonly used protocols for the detection of UV mutagenesis are not optimal for this strain. Indeed, when bacteria were exposed to a UV-C irradiation dose of 5 J/m2 (this dose was about 20-fold lower than the commonly used dose), the strong UV mutagenesis phenotype was detected also with the strain PC20 (Fig 6 and S9 Table). Under such conditions the Rifr mutant frequency increased about 60 times. Interestingly, the irradiation of cells at UV-C irradiation at a dose of 5 J/m2 elevated remarkably the UV mutagenesis also in the case of the other three studied strains PC24, PaWrulAB and PaW1 (Fig 6 and S9 Table). For example, the strain PC24 which exhibited about 4.5-fold increase in the number of Rifr mutants above spontaneously induced levels at a dose of 100 J/m2 had approximately 38-fold elevated mutant frequency when cells were exposed to the UV irradiation at a dose of 5 J/m2. Even the reference strain P. putida PaW85 exhibited a clearly visible UV mutagenesis phenotype at this dose (4.8-fold elevated mutant frequency above the spontaneous levels). Thus, without testing for a wider range of environmental conditions one may easily underestimate a potential of the studied strains for the induced mutagenesis.
Is the degradation of aromatic compounds mutagenic for bacteria?
Exposure of bacteria to aromatic compounds induces production of reactive oxygen species (ROS) which may cause DNA damage. This in turn may cause elevated mutation frequency and speed up the evolution of novel metabolic properties. To examine possible mutagenic effects of phenolic compounds, we incubated bacteria in the presence of phenol and 3-methylphenol (m-cresol). Notably, the indigenous strains examined in the current study are able to degrade various aromatic compounds (Table 1). According to the previously published data the strain PC20 can grow on phenol, p-cresol, salicylate and naphthalene, and the strain PC24 can grow on phenol, p-cresol and m-toluate, whereas PC20 degrades p-cresol via meta pathway and PC24 via protocatechuate pathway [24, 44, 45]. Here it is important to mention that during the current studies we discovered that the strain PC20 cannot grow any more on minimal medium containing p-cresol as a carbon source (data not shown). Importantly, both PC20 and PC24 contained the mono-component phenol monooxygenase gene pheA probably acquired after the release of laboratory bacteria employed for removal of phenolic pollution in contaminated water [24, 46]. The latter inspired us to use the phenol-degrading P. putida strain PaWpheBA25 with the introduced pheBA operon as a control. P. putida strain PaW85 unable to degrade phenol was also included for the comparison. Bacteria were grown on glucose CAA media similarly to that when we measured the spontaneous mutation frequency, except phenol or m-cresol was also added.
To our surprise, all of the studied strains had similar or even reduced Rifr mutant frequencies when phenol was present in the growth medium (Fig 7 and S10 Table). In the case of the strain PC20 the addition of m-cresol also lowered the Rifr mutant frequency. Presented are only these results obtained with 5 mM phenol and with 2.5 mM m-cresol. Lower concentrations of these compounds had no effect on mutant frequency (data not shown). The results of this experiment implied that the studied bacteria have evolved highly functional mechanisms to cope with external or internal oxidative stress in the presence of phenolic compounds when the environment contains energetically rich growth substrates (glucose and amino acids). Therefore, in order to examine whether the availability of carbon and energy sources could affect potential mutagenic effects of the studied phenolic compounds, we repeated the experiment by cultivating bacteria in M9 minimal medium which either contained glucose, phenol or p-cresol as an only carbon source. No amino acids were supplied for the growth medium (for more details of the growth conditions of bacteria, see the Materials and Methods section). Indeed, it turned out that the growth of bacteria on minimal medium revealed the mutagenicity of phenol (Fig 8 and S11 Table). However, this appeared only under certain circumstances. Namely, the strain PaWpheBA25 (obtained by the introduction of the phenol degradation operon pheBA into the strain P. putida PaW85) which was constructed in the laboratory immediately before its use in the mutagenesis assay exhibited on phenol minimal medium about 4.2-fold elevated Rifr mutant frequency in comparison to the growth on glucose minimal medium (Fig 8). At the same time, the indigenous isolate PC24 exhibited comparable frequency of Rifr mutants on both growth substrates, and the strain PC20 exhibited even ~ 2.7 times reduced Rifr mutant frequency when grown on phenol in comparison with the growth on glucose minimal medium. Thus, taken together, our results indicated that the mutagenic effect of aromatic pollutants depends on the availability of other carbon sources in the environment. These results also suggested that mutagenicity could be affected by a factor of how long bacteria have evolved to use a particular pollutant as a carbon source. Further studies are needed to test this hypothesis.
Phenol or m-cresol was supplemented into M9 glucose medium supplemented with CAA for the strains PaW85, PaWpheBA, PaWrulAB and PC24 at the following concentrations: phenol (5 mM), m-cresol (2.5 mM). These compounds were added in lower concentrations (2.5 mM phenol and 1.25 mM m-cresol) for the strain PC20 because of growth inhibiting effects at higher concentrations. Data were analysed using the Kruskal-Wallis test (S10 Table). *** designates P<0.0001.
The frequency of mutations is usually held at low level. However, although most mutations are deleterious, bacteria with elevated spontaneous mutation frequency have been isolated from populations of human pathogens [16, 17, 47]. One of the strongest mutator phenotypes is caused by the inactivation of DNA mismatch repair (MMR) gene mutS which leads up to 100-fold higher mutation frequency [10, 48]. Populations can be enriched with mutators via several population bottlenecks occurring after strong selective challenges. For example, P. aeruginosa populations chronically infecting the lungs of CF-patients have to adapt to the highly compartmentalized and anatomically deteriorating lung environment, as well as to the challenges of the immune defence and antibiotic therapy [16, 18].
Is it possible that the evolution of bacteria with the capacity to degrade toxic environmental pollutants is also accompanied with the appearance of constitutive mutators? It is known that above a certain threshold organic pollutants affect survival of bacteria by altering the cell membrane structure or by impairing biosynthetic pathways essential for bacterial growth [20, 21, 49]. In addition, incomplete degradation of such compounds by bacteria lacking entire functional pathways for their catabolism may lead to the formation of toxic dead-end metabolites. Thus, bacteria which are evolving to degrade pollutants might be faced with strong selective pressure to survive in the presence of these toxicants. In the current study we assessed the existence of constitutive mutators within 53 indigenous isolates, mostly pseudomonads, which can degrade various toxic pollutants (e.g., phenol, toluene, and their methyl derivatives; alkanes) as a carbon source. Notably, the strains examined for a mutator phenotype in the current study included a number of pseudomonads which were isolated six years after an accidental fire in the oil shale mine in Estonia in 1988 [24, 46]. To reduce the produced pollution in situ, phenol-degrading P. putida laboratory strains carrying the phenol degradation operon pheBA on plasmid were deliberately released in the contaminated water. Several indigenous isolates examined in the current study have acquired this operon by the horizontal gene transfer [24, 46]. Thus, at least some of the investigated strains carry a signature that they have evolved under the conditions of heavily polluted environments. However, we did not find any strong mutators among the examined bacterial strains. Only two of the investigated strains expressed a moderate mutator phenotype (Fig 1). These results implied that constitutively increased mutability has not been prevalent in the evolution of bacteria exposed to toxic environmental pollutants. However, it should be noted that constitutive mutators might arise and persist only on a limited basis [50, 51]. In vivo experiments have shown a potential benefit of hypermutation for bacterial adaptation to new environments, but once adapted, this advantage disappears and the transmissibility of the mutator strains is considerably reduced . When passed through severe bottlenecks, mutator lineages also accumulate mutations that confer loss in fitness [52, 53]. We suppose that the growth of pathogens in a body of its host might be much more limited than the growth of environmental bacteria living in a soil or water reservoirs. Once adapted to consume a toxic pollutant, environmental bacteria could have passed many more generations in open environment than pathogens in their host. Therefore, we cannot exclude a possibility that some of the isolated indigenous strains have lost their mutator phenotype. Drawing parallels from the literature, the results of the mutS gene sequence analysis from natural E. coli isolates have indicated that MMR functions have been repeatedly lost and reacquired during the evolutionary history of E. coli [54, 55]. Also, the experimental evidence of evolution towards reduced mutation rates in a ΔmutS mutator E. coli population was recently demonstrated .
About 11% of the investigated indigenous strains were hypomutators (Fig 1). It should be noted that hypomutable isolates have also been identified in several other bacterial populations, e.g., among clinical and environmental isolates of P. aeruginosa, E. coli, and Stenotrophomonas maltophilia [57–60]. The presence of hypomutators (or antimutators) along with isolates exhibiting increased mutation frequency indicates that mutator and antimutator alleles can often arise and spread in natural bacterial populations. Evolutionary forces affecting the genomic mutation rate by natural selection and random genetic drift have been extensively discussed, e.g., in [61–64].
The frequency of mutations can also be temporally induced. For example, mistranslation, and accumulation of oxidative and alkylation damage may transiently increase mutagenesis [10, 65, 66]. In addition, under stressful conditions such as exposure to antibiotics or starvation, DNA synthesis may become more erroneous because of the induction of error-prone DNA polymerases, resulting in a situation in which DNA repair systems are unable to cope with increasing amounts of replication errors [67–69]. Several bacterial DNA polymerases are induced as a part of SOS regulon in response to DNA damage. These DNA polymerases perform translesion DNA synthesis (TLS) when replication forks have collapsed at a blocking lesion whereas the DNA synthesis may be error-prone resulting in increased frequency of mutations in bacteria [11, 12, 70]. The results of our work revealed that a large proportion of the studied indigenous strains exhibit UV-irradiation-induced mutagenesis (Fig 2 and S6 Table). In many bacteria the UV-mutagenesis occurs as a consequence of TLS either performed by DNA polymerase Pol V or ImuC (former DnaE2) . Error-prone DNA polymerase Pol V homologues are frequently encoded by naturally occurring catabolic conjugative plasmids and may facilitate generation of metabolic diversity in bacteria [35, 71, 72]. A closer inspection of two indigenous strains P. fluorescens PC20 and PC24 revealed that both strains harbour genes for ImuC and more than one copies of genes for Pol V.
The transcription of the genes encoding for Pol V and ImuC was induced under the conditions of DNA damage (Fig 5A). However, the level of the induction of the transcription of the rulAB genes by MMC was lower in the strains PC20 and PC24 than that of the rulABpWW0 in the positive control strain PaW1 (Fig 5A). At the same time, the rulAB operons of PC20 and PC24 were expressed at higher basal level in comparison with the rulABpWW0 operon (Fig 5B). The transcription of the DNA damage-inducible genes is negatively regulated by the LexA repressor protein which in E. coli binds to the conserved sequences CTGT-N8-ACAG . Functional E. coli LexA homologues have been found in many other bacterial species as well [reviewed e.g., in ]. For instance, we have previously shown that the promoter region of the rulAB genes on pWW0 contains a DNA sequence showing a perfect match to the E. coli LexA-binding consensus sequence . Analysis of the putative promoter sequences of the rulAB genes in the strains PC20 and PC24 also revealed the presence of a sequence which matched completely to the E. coli LexA repressor-binding consensus (S5 Fig). However, although there is a minimal LexA-binding consensus sequence, the timing and the level of induction can vary for each LexA-regulated gene because of multiple factors. The nucleotides surrounding the LexA consensus can influence the LexA binding affinity, and the strength of the promoter and the location of the LexA boxes relative to the promoter could also have an impact on the induction level [75, 76]. In addition, it should be noted that although the binding sites for the E. coli-like LexA repressor are similar for E. coli and Pseudomonas species, there might be some differences in the binding affinity. This assumption is supported by our previously published results demonstrating that the promoters of the rulABpWW0 and the promoter of the DNA polymerase Pol IV gene dinB of P. putida carrying the E. coli-like LexA box are induced at higher level in E. coli than in P. putida [35, 77]. Bacteria living in soil and water reservoirs are constantly confronted with stressful conditions such as starvation for nutrients, desiccation, and non-optimal pH and temperature, all of which can induce DNA damage. Thus, one may speculate that soil bacteria, e.g., pseudomonads, have evolved a regulatory system enabling higher basal level of transcription of error-prone DNA polymerase genes because they have adapted to live in harsh and fluctuating environmental conditions inducing frequent injury in DNA.
Analysis of the putative promoter regions of the imuC-containing operon from PC20 and PC24 revealed the presence of the LexA2 binding sequences GTAC-N4-GTRC upstream of the structural genes (S6 Fig). Treatment of bacteria by DNA damaging agent MMC induced transcription of the imuC (Fig 5) which is in good accordance with the imuC expression profile in the other bacterial species [41–43]. Interestingly, although the strains PC20 and PC24 possessed several copies of DNA damage-induced DNA polymerase genes for Pol V and ImuC, according to the results of the measurement of UV tolerance both strains classified to a group exhibiting reduced tolerance to the UV irradiation in comparison with the strains PaWrulAB and PaW1 (Table 2 and S5 Table). The strain PC24 was even more sensitive to the UV irradiation than the rulAB-deficient reference strain PaW85 (Table 2). Besides TLS, the tolerance against UV irradiation can be affected by several other mechanisms. The main mechanism to tolerate UV-damage is nucleotide excision repair pathway (NER) which repairs the damage caused by UV irradiation in non-mutagenic way [78, 79]. However, analysis of the draft genome sequence of PC20 and PC24 revealed the presence of all NER pathway genes uvrA, uvrB, uvrC and uvrD (S7 Fig).
Here we wish to note that the role of ImuC/DnaE2 as an error-prone TLS polymerase protecting cells against DNA damage may not be general. For example, Streptomyces coelicolor dnaE2 was found to be SOS-inducible, but it was not involved in UV tolerance or mutagenesis . Similarly, we have previously reported that the deletion of the dnaE2 (imuC) gene from the chromosome of P. putida PaW85 did not affect cells’ tolerance to the UV irradiation . At the same time, the role of P. putida ImuC as a mutagenic DNA polymerase seems to be controversial. We have observed that the absence of this DNA polymerase increases the frequency of accumulation of mutations in carbon-starved P. putida . On the other hand, the presence of ImuC altered the spectrum of mutations in P. putida only in the case when bacteria were exposed to UV-C irradiation (no differences could be detected without treatment of cells by DNA damaging agents), suggesting its participation in DNA replication under the conditions of DNA damage . Moreover, our recent data  have indicated that mutagenesis induced by DNA alkylating agent MMS is largely ImuC-dependent both in P. putida and P. aeruginosa. Thus, participation of this DNA polymerase in mutagenic pathways could vary depending on the environmental conditions of bacteria.
Mechanisms reducing mutation rates (e.g., proofreading during DNA replication, expression of DNA repair functions, export or degradation of mutagenic compounds) could be energetically costly for cells. Therefore, it is reasonable to assume that rich growth environment allows more resources for establishment of DNA integrity. If the amount of nutrients becomes limiting, bacteria could start saving energy for other cellular functions by shutting down replication fidelity mechanisms. For instance, there is evidence that starvation is accompanied by increased frequency of mutations [67, 68, 84, 85]. The results of the current study implied that the availability of nutrients may affect frequency of mutations in bacteria (compare the results in Figs 7 and 8). Specifically, when we examined the effect of pollutants (e.g., phenol) on the induction of Rifr mutations, we did not observe any increasing effects of phenol on the mutation frequency if bacteria were grown in the environment supplemented by glucose and amino acids. The mutation frequency was increased only in minimal medium when phenol was served as an only carbon and energy source (Fig 8).
Nevertheless, the mutagenic effect of phenol appeared only with the strain PaWpheBA which was constructed by introducing the phenol degradation genes pheBA into the reference strain P. putida PaW85. On the contrary, the mutation frequency of the indigenous strains PC20 and PC24 was declined on phenol minimal medium. As already mentioned above, the strains PC20 and PC24 have most likely acquired the pheBA operon from laboratory strains released into contaminated water after fire accident in 1988 . Indigenous bacteria present in the area of phenolic contamination possibly met high concentration of the pollutants both before and after the acquisition of the pheBA operon. Bacterial strains obtained the pheBA operon were isolated 6 years later, which have allowed mutual adaptation between the bacterial host and the acquired pathway genes by thousands of generations. Phenol like other environmental pollutants act both as nutrients and as a stressors for bacteria . Therefore, efficient phenol degraders should evolve also efficient mechanisms to deal with phenolic stress. Hence, one may speculate that while degrading phenolic compounds, the indigenous strains PC20 and PC24 have developed stress response mechanisms which can reduce mutational load when exposed to these pollutants. Further studies are in progress to test this idea.
The evolution of environmental pollutants degrading bacteria is an ongoing process. Both horizontal transfer of genes and mutagenesis induced under stressful conditions might facilitate development of new catabolic pathways. It has been suggested that growth of bacteria in the presence of pollutants is also mutagenic. The indigenous bacterial isolates characterized in the current study, many of them collected from polluted environments, did not display a strong constitutive mutator phenotype. Nevertheless, a large proportion of the investigated strains exhibited a potential for induced mutagenesis under the conditions of DNA damage. Two phenol and cresol degrading P. fluorescens strains selected for a deeper characterization possessed several copies of DNA damage-induced error-prone DNA polymerases Pol V and ImuC. However, it is possible that compared to the laboratory-constructed strains these indigenous bacteria have already evolved an efficient mechanism for degradation of these phenolic compounds without increasing mutation frequency. Further elucidation of mechanisms of stress response in these indigenous strains would be useful to understand processes of efficient bioremediation. Also, laboratory experimental evolution experiments with bacteria which have acquired novel catabolic genes for the degradation of pollutants only lately would shed light in the evolution of new metabolic properties of bacteria.
Materials and methods
Bacterial strains and media
The bacterial strains used in this study are described in Table 1. All indigenous strains listed in Table 1 are deposited in the Collection of Environmental and Laboratory Microbial Strains (CELMS; available in the Estonian Electronic Microbial dataBase (EEMB) website http://eemb.ut.ee). Complete medium used was lysogeny broth (LB) medium  or R2A , and minimal medium was M9 (Adams, 1959). Solid medium contained 1.5% Difco agar. Casamino acids (CAA) and glucose were added to the minimal medium at final concentrations of 0.4% and 0.2%, respectively. Aromatic compounds were added to the media in following concentrations: phenol (1.25 mM, 2.5 mM or 5mM); m-cresol (1.25 mM, 2.5 mM) and p-cresol (1.25 mM). Rif-LB plates contained 100 mg/ml of rifampicin (Rif) and Sm-LB plates contained 200μg/ml of streptomycin (Sm). All strains were incubated at 30°C and liquid cultures were grown on rotary shakers (180 rpm).
Estimation of spontaneous mutation frequency by fluctuation test
In order to estimate spontaneous mutation frequency, we performed the fluctuation tests and calculated the median value for mutants per 1 x 109 cells as described in . The frequency of Rifr mutants was determined at least in 13 independent cultures for each studied bacterial strain. Independent cultures were generated by growing cells to late logarithmic growth phase in M9 medium which contained glucose (0.2%) and CAA (0.4%). The cultures were then diluted by 105 into fresh glucose and CAA-containing M9 medium. After that 2.3-ml aliquots were dispensed into test tubes and cells were allowed to grow 24 h to reach saturation. Approximately 2 × 109 cells were plated onto Rif-LB plates. Emergence of Rifr colonies were counted after 72 h of incubation. The same protocol was also used to study the frequency of streptomycin resistant mutants (Smr) with an exception of plating cells to Sm-LB plates (Sm 200 μg/ml). To exclude possibility that the estimated mutant frequencies of the studied environmental isolates were affected by variations in the tolerance to Rif or Sm, we measured the minimal inhibitory concentrations (MIC) of used antibiotics on a selection of strains (10 strains). For MIC assay the cells were grown overnight in glucose and CAA-containing M9 medium and then diluted 100 fold into fresh medium containing serial dilutions of antibiotics. Cells were incubated in microtiter plates at 30°C for 24h and then culture density was measured at OD580. MIC was defined as the lowest concentration of antibiotic which prohibit bacterial growth.
Estimation of the effect of aromatic compounds on the frequency of Rifr mutants
The influence of aromatic compounds in the growth medium was studied in high and low nutrient M9 medium. In the case of high nutrient medium independent cultures were generated by growing cells overnight in M9 medium supplemented with glucose and CAA. The cultures were then diluted by 105 into fresh glucose and CAA containing M9 medium without adding any other compounds or supplemented with either 2.5 mM or 5 mM phenol, or 1.25 mM or 2.5 mM m-cresol. Next 2.3 ml aliquots were dispensed into flasks and cells were allowed to grow 48 h to reach saturation. Approximately 2 × 109 cells were plated onto Rif-LB plates. Emergence of Rifr colonies was counted after 72 h of incubation. In the case of low nutrient medium independent cultures were generated by growing cells overnight in M9 medium supplemented with glucose. The cultures were then diluted by 100 fold into fresh M9 minimal medium containing either glucose, phenol (2.5 mM) or p-cresol (1.25 mM) as the only carbon source. Next 15 ml aliquots were dispensed into flasks and cells were allowed to grow 48 h to reach saturation. Approximately 2 × 109 cells were plated onto Rif-LB plates. The emergence of Rifr colonies were counted after 72 h of incubation. The median value for Rifr mutants was calculated as described above.
UV-irradiation survival studies
The studied strains were grown overnight in glucose and CAA containing M9 media, and the serial of dilutions, 10−4 to 10−9 from the initial overnight culture, were placed as 10 μl drops on the LB agarose plates. The plates with dilutions were then subjected to UV-C irradiation which was performed at a 254-nm wavelength at a dosage of 10, 20, or 40 J/m2 using a CX—2000 Crosslinker. Irradiated plates were incubated in the dark for 24 h before the enumeration of CFU. Survival was expressed as the number of CFU detected after irradiation as a percentage of those detected after no treatment. At least 3 independent experiments were performed for each strain.
To ensure the exponential growth of bacteria, the cultures grown overnight in glucose and CAA containing M9 medium were diluted 1:100 into fresh medium. After the cultures reached mid-exponential phase (optical density at 580 nm of 0.6 to 0.8), 11 ml of each culture was transferred to a 100-mm-diameter sterile Petri dish for UV-irradiation. The irradiation with UV-C light was performed at a 254-nm wavelength at a dose of 100 J/m2 (if not stated otherwise in the Results section) using a CX—2000 Crosslinker. Aliquots (1.6 ml) were removed from the Petri dish and grown in test tubes overnight with shaking at 180 rpm. 500 μl of overnight culture was plated onto rifampicin plates and the appearance of mutants was analysed as described before. However, if the amount of Rifr colonies was uncountable on the plates, the assay was repeated and only 100 μl of overnight culture was plated.
Nucleotide sequencing and in silico analysis
The total genomic DNA from the bacterial strains PC20 and PC24 was isolated and the whole genome sequencing was performed as described in  (S1 Appendix). In case of PC20, the total length of the assembled contigs was 7,856,954 bp (230-fold sequencing coverage), in 3749 contigs. The contigs N50 was 52,853 bp with the longest scaffold 204,810 bp. In case of PC24, the total length of the assembled contigs was 6,457,994 bp (251-fold sequencing coverage), in 878 contigs. The contigs N50 was 101,312 bp with the longest scaffold 259,869 bp. Gaps in the contigs used in this study were sequenced with 3730xl DNA Analyzer (Applied Biosystems) using the BigDye® Terminator v3.1 Cycle Sequencing Kit (Applied Biosystems) and protocols provided by the manufacturer.
The contigs of PC20 and PC24 containing the genes involved in DNA replication and repair were detected from all obtained contigs using the program BLASTP version 2.2.31 with default parameters except for the E value cutoff which was 10−3. The database file for searches was created from sequences obtained from NCBI’s Protein Clusters database (1976 sequences of different proteins involved in DNA replication and repair). The ORFs of the selected contigs were subsequently predicted using the program Prodigal version 2.6.1 . Sequences were aligned and phylogenetic trees were constructed with ClustalX version 2.1  and further visualised with TreeViewX version 0.5.0 .
The nucleotide sequences analysed in the current study are deposited in GenBank (see S7 Table for GenBank accession numbers).
RNA extraction and RT-qPCR
In order to study the expression rate of error-prone DNA polymerases, the cultures were grown overnight in glucose and CAA containing M9 medium and then diluted 1:100 into fresh medium. After two hours of growth, the expression of DNA damage tolerance mechanisms was induced for 2 h with the addition of MMC mitomycin C (MMC) with the final concentration of 2μg/ml. Next the RNA was extracted using RNAzol®RT kit (Molecular Research Centre, Inc) with manufacturer’s instructions from normal and MMC induced cells. Additional treatment with DNase I (Fermentas) was performed according to the protocol (Fermentas) after which the quantity of RNA was measured with spectrophotometer (The NanoDrop™ 1000, Thermo Scientific). The quality of RNA was first inspected by electrophoresis (1% agarose) for the integrity and secondly with PCR for contamination with DNA.
The quantitative reverse transcription-PCR (qRT-PCR) assay was performed with SuperScript® III One-Step RT-PCR System with Platinum® Taq DNA Polymerase (Invitrogen) in the total volume of 10 μl. The RT-qPCR was operated in the real-time PCR system Rotor-Gene® Q (Qiagen). In RT-qPCR the primers were used at concentration of 0.4 μM and their sequences are presented in S12 Table. The primers were designed to generate products of ~200 bp. The reaction conditions were: 50°C for 3 min, 95°C for 5 min, followed by 40 cycles of denaturation at 95°C for 15 s, annealing at 58°C for 30 s, and extension at 72°C for 20 s. The fluorescence intensity of SYBR Green was measured automatically at the end of the extension step. At the end of run, melting curve analyses was performed increasing the temperature from 72°C to 95°C, using 4 s and 1°C interval with continuous fluorescence recording. Raw data from the RT-qPCR was analyzed with Rotor-Gene Series software, version 188.8.131.52 (Qiagen) and mRNA quantities were calculated using the LinRegPCR software version 2015.3 . Data from at least three separate RT-qPCR experiments were averaged and normalized against the trpA levels, which expression is not affected by DNA damage .
Statistical analysis of the results
Sharipo-Wilk test for normality was used for all distributions and because normality was not addressed, the nonparametric Mann-Whitney U test or Kruskal-Wallis test was used for comparisons. The calculations were performed using Statistica 10 software [StatSoft, Inc. (2011). STATISTICA (data analysis software system), version 10. www.statsoft.com.]. For statistical tests the significance level was set at P = 0.05. In order to reduce the amount of potential errors on multiple comparisons, Benjamini-Hochberg procedure was used with false discovery rate (FDR) of 0.05 .
Original datasets used to compute statistics are available as a part of supporting information (S13 Table).
S1 Appendix. Experimental procedure for whole genome sequencing.
S1 Fig. Comparison of Smr and Rifr resistant mutant frequencies in representatives of the indigenous strains and the reference strain P. putida PaW85.
The median value of the frequency of Smr mutants of the strains C70 and P4 was 0 (the mean values were 1.65 x 10−9 and 4.8 x 10−10, respectively).
S2 Fig. Multiple sequence alignment of RulA homologs from Pseudomonas fluorescens PC20, PC24; Pseudomonas putida plasmid pWW0 RulA (Q8VMP5) and E. coli K-12 UmuD (P0AG11).
Black boxes indicate the putative active sites residues according to NCBI Conserved Protein Domain Family database (source pfam; superfamily cl10465; Peptidase_S24_S26; . Residues shaded orange represent the autocatalytic cleavage site shown to be required for posttranslational processing of UmuD/RulA homologs [96, 97]. Residues surrounded by pink boxes have been shown to be important in the UmuDC mediated mutagenesis . Sequences were aligned with ClustalX2 and domains were marked according to Protein sequence analysis and classification portal Interpro (http://www.ebi.ac.uk/interpro/) based on the sequence of UmuD (P0AG11).
S3 Fig. Multiple sequence alignment of RulB homologs from Pseudomonas fluorescens PC20, PC24; Pseudomonas putida plasmid pWW0 RulB (Q8VMP6) and E. coli K-12 UmuC (P04152).
Black boxes indicate the putative active site residues according to NCBI Conserved Protein Domain Family database (source cd00424; superfamily cl12025; PolY_Pol_V_umuC; . Residues surrounded by pink boxes have been shown to be important in the UmuDC mediated mutagenesis [98–106]. Sequences were aligned with ClustalX2 and domains were marked according to Protein sequence analysis and classification portal Interpro (http://www.ebi.ac.uk/interpro/) based on the sequence of UmuC (P04152).
S4 Fig. Multiple sequence alignment of ImuC/DnaE2 homologs from Pseudomonas fluorescens PC20.
PC24, Pseudomonas putida KT2440 DnaE2/ImuC (Q88182), and Mycobacterium tuberculosis (A0A089QXJ1). Black boxes indicate the putative active site residues according to NCBI Conserved Protein Domain Family database (source cd07431; superfamily cl23724; PHP_PolIIIA_DnaE2; . Residues shaded gray represent active-site amino acids which are highly conserved among Y-family polymerases . Residues underlined with red are conserved active site regions according to the study by . Sequences were aligned with ClustalX2 and domains were marked according to Protein sequence analysis and classification portal Interpro (http://www.ebi.ac.uk/interpro/) based on the sequence of P. putida KT2440 DnaE2/ImuC (Q88182). Region underlined with black is a C-terminal amino acid motif -[S/T/G]R[D/N]F[D/R/H]- highly conserved in DnaE2-type proteins .
S5 Fig. Multiple sequence alignment of putative promoter regions of rulAB genes and E. coli umuDC genes.
The -35 and -10 hexamers of the promoters are marked by black boxes and transcriptional start site for umuDC promoter  is marked with red box. LexA-binding consensus sequence is aligned on the last row . Sequences were aligned with ClustalX2.
S6 Fig. Multiple sequence alignment of putative promoter regions of imuABC.
The -35 and -10 hexamers of the promoters are marked by black boxes. LexA2-binding consensus sequence is aligned on the last row . Sequences were aligned with ClustalX2.
The phylogenetic trees of UvrA (A), UvrB (B), UvrC (C) and UvrD (D) protein sequences. Sequences were aligned with ClustalX2 and further visualised with TreeViewX. Numbers at branch nodes indicate bootstrapping values for 1000 bootstrap replicates. Values under 500 were removed. The identifiers of the aligned sequences are presented in S7 Table.
S1 Table. The median values of the frequency of spontaneous Rifr mutants.
Three laboratory strains were studied as a comparison in addition to the environmental strains. All of the mutant frequencies were compared to P. putida PaW85 laboratory reference strain with Kruskal-Wallis test. The statistically significant p-values according to Benjamini-Hochberg procedure are indicated with red (FRD = 0.05).
S2 Table. The minimal inhibitory concentrations of representatives of the indigenous strains and the reference strain P. putida PaW85.
S3 Table. The results of Mann-Whitney U test for comparing the appearance frequency of mutants for two different antibiotics (rifampicin and streptomycin) within one strain.
The statistically significant p-values according to Benjamini-Hochberg procedure are indicated with red (FRD = 0.05).
S4 Table. Comparison of Smr mutant frequencies with Kruskal-Wallis test against PaW85.
The statistically significant p-values according to Benjamini-Hochberg procedure are indicated with red (FRD = 0.05).
S5 Table. The survival percentage of bacteria after UV-C irradiation.
Underlined are the strains with at least 10 fold higher UV-tolerance (UV- 20 J/m2) compared to PaW85. Kruskal-Wallis test was performed to compare UV-tolerance (UV- 20 J/m2) of the studied strains against PaW1, which possesses PolV coding operon rulAB. The statistically significant p-values according to Benjamini-Hochberg procedure are indicated with red (FRD = 0.05).
S6 Table. The UV-induced (100 J/m2) Rifr mutant frequencies.
The Mann-Whitney U test was performed to distinguish statistically significant differences of UV-induced and spontaneous Rifr mutant frequencies. The statistically significant p-values according to Benjamini-Hochberg procedure are indicated with red (FRD = 0.05). The difference between spontaneous and UV-induced mutant frequency is presented as fold of induction. Underlined are the strains with at least 10 fold higher tolerance against UV irradiation (UV-20 J/m2) compared to PaW85.
S7 Table. The sequences used in phylogenetic analyses and multiple sequence alignments.
Protein sequences were translated based on the gene sequences listed below. Sequence ID-s were obtained from GenBank (https://www.ncbi.nlm.nih.gov/genbank/) and The European Nucleotide Archive (ENA, http://www.ebi.ac.uk/ena).
S8 Table. The statistical analysis for distinguishing significant changes in the expression rate of the genes of error-prone DNA polymerases after treatment with MMC.
The RT-qPCR results were analyzed with Mann-Whitney U test and p-values indicated with red represent statistically significant differences according to Benjamini-Hochberg procedure (FRD = 0.05).
S9 Table. The comparison of spontaneous and UV-induced (5 J/m2) mutant frequencies.
The Mann-Whitney U test was performed to distinguish statistically significant differences of UV-induced and spontaneous Rifr mutant frequencies according to Benjamini-Hochberg procedure (FRD = 0.05). Red indicates the statistically significant differences. The difference between spontaneous and UV-induced mutant frequency is presented as a fold of induction.
S10 Table. The effect of aromatic compounds (2.5 mM m-cresol or 5 mM phenol with exception of PC20 for which we used 2.5 mM phenol and 1.25 mM m-cresol) to the appearance frequency of Rifr mutants in rich medium (M9 medium supplemented with glucose (0.2%) and CAA (0.2%)).
Multiple comparisons with Kruskal-Wallis test were performed to distinguish statistically significant differences in the appearance frequency of Rifr mutants in the presence of aromatic compounds. Red indicates the statistically significant differences according to Benjamini-Hochberg procedure (FRD = 0.05). The effect of aromatic substrates to the frequency of Rifr mutants is presented as a fold of induction compared to the cells grown on medium without aromatic compounds.
S11 Table. The frequency of Rifr mutants grown in M9 minimal medium supplemented with either glucose (0.2%), phenol (2.5 mM) or p-cresol (1.25 mM) as a sole carbon and energy source.
The Mann-Whitney U test was performed to distinguish statistically significant differences in the appearance frequency of Rifr mutants in the presence of different carbon sources. Red indicates the statistically significant results according to Benjamini-Hochberg procedure (FRD = 0.05). The effect of aromatic substrates on the frequency of Rifr mutants is presented as a fold of induction compared to the cells grown on glucose as a sole carbon source.
S12 Table. Primers used in the RT-qPCR experiment.
We would like to thank Heili Ilves, Katren Mikkel, Mari Tagel and Kärt Ukkivi for the critical reading of the manuscript. We are also very grateful to Eeva Heinaru for sharing her immeasurable experience in working with indigenous bacterial strains.
- 1. Juhas M, van der Meer JR, Gaillard M, Harding RM, Hood DW, Crook DW. Genomic islands: tools of bacterial horizontal gene transfer and evolution. FEMS Microbiol Rev. 2009;33(2):376–93. pmid:19178566
- 2. Kivisaar M. Degradation of nitroaromatic compounds: a model to study evolution of metabolic pathways. Mol Microbiol. 2009;74(4):777–81. pmid:19818019
- 3. Nojiri H, Shintani M, Omori T. Divergence of mobile genetic elements involved in the distribution of xenobiotic-catabolic capacity. Appl Microbiol Biotechnol. 2004;64(2):154–74. pmid:14689248
- 4. Norman A, Hansen LH, Sorensen SJ. Conjugative plasmids: vessels of the communal gene pool. Philos Trans R Soc Lond B Biol Sci. 2009;364(1527):2275–89. pmid:19571247
- 5. van der Meer JR, Sentchilo V. Genomic islands and the evolution of catabolic pathways in bacteria. Curr Opin Biotechnol. 2003;14(3):248–54. pmid:12849776
- 6. Perez-Pantoja D, Nikel PI, Chavarria M, de Lorenzo V. Endogenous stress caused by faulty oxidation reactions fosters evolution of 2,4-dinitrotoluene-degrading bacteria. PLoS Genet. 2013;9(8):e1003764. pmid:24009532
- 7. Drake JW. A constant rate of spontaneous mutation in DNA-based microbes. Proc Natl Acad Sci U S A. 1991;88(16):7160–4. pmid:1831267
- 8. Sniegowski PD, Gerrish PJ, Lenski RE. Evolution of high mutation rates in experimental populations of E. coli. Nature. 1997;387(6634):703–5. pmid:9192894
- 9. MacLean RC, Torres-Barcelo C, Moxon R. Evaluating evolutionary models of stress-induced mutagenesis in bacteria. Nat Rev Genet. 2013;14(3):221–7. pmid:23400102
- 10. Miller JH. Spontaneous mutators in bacteria: insights into pathways of mutagenesis and repair. Annu Rev Microbiol. 1996;50:625–43. pmid:8905093
- 11. Nohmi T. Environmental stress and lesion-bypass DNA polymerases. Annu Rev Microbiol. 2006;60:231–53. pmid:16719715
- 12. Ippoliti PJ, Delateur NA, Jones KM, Beuning PJ. Multiple strategies for translesion synthesis in bacteria. Cells. 2012;1(4):799–831. pmid:24710531
- 13. Michaels ML, Cruz C, Grollman AP, Miller JH. Evidence that MutY and MutM combine to prevent mutations by an oxidatively damaged form of guanine in DNA. Proc Natl Acad Sci U S A. 1992;89(15):7022–5. pmid:1495996
- 14. LeClerc JE, Li B, Payne WL, Cebula TA. High mutation frequencies among Escherichia coli and Salmonella pathogens. Science. 1996;274(5290):1208–11. pmid:8895473
- 15. Matic I, Radman M, Taddei F, Picard B, Doit C, Bingen E, et al. Highly variable mutation rates in commensal and pathogenic Escherichia coli. Science. 1997;277(5333):1833–4. pmid:9324769
- 16. Oliver A, Canton R, Campo P, Baquero F, Blazquez J. High frequency of hypermutable Pseudomonas aeruginosa in cystic fibrosis lung infection. Science. 2000;288(5469):1251–4. pmid:10818002
- 17. Hall LM, Henderson-Begg SK. Hypermutable bacteria isolated from humans—a critical analysis. Microbiology. 2006;152(Pt 9):2505–14. pmid:16946246
- 18. Sundin GW, Weigand MR. The microbiology of mutability. FEMS Microbiol Lett. 2007;277(1):11–20. pmid:17714481
- 19. Taddei F, Radman M, Maynard-Smith J, Toupance B, Gouyon PH, Godelle B. Role of mutator alleles in adaptive evolution. Nature. 1997;387(6634):700–2. pmid:9192893
- 20. Cases I, de Lorenzo V. Promoters in the environment: transcriptional regulation in its natural context. Nat Rev Microbiol. 2005;3(2):105–18. pmid:15685222
- 21. Ramos JL, Krell T, Daniels C, Segura A, Duque E. Responses of Pseudomonas to small toxic molecules by a mosaic of domains. Curr Opin Microbiol. 2009;12(2):215–20. pmid:19269884
- 22. Dominguez-Cuevas P, Gonzalez-Pastor JE, Marques S, Ramos JL, de Lorenzo V. Transcriptional tradeoff between metabolic and stress-response programs in Pseudomonas putida KT2440 cells exposed to toluene. J Biol Chem. 2006;281(17):11981–91. pmid:16495222
- 23. Santos PM, Benndorf D, Sa-Correia I. Insights into Pseudomonas putida KT2440 response to phenol-induced stress by quantitative proteomics. Proteomics. 2004;4(9):2640–52. pmid:15352239
- 24. Heinaru E, Truu J, Stottmeister U, Heinaru A. Three types of phenol and p-cresol catabolism in phenol- and p-cresol-degrading bacteria isolated from river water continuously polluted with phenolic compounds. FEMS Microbiol Ecol. 2000;31(3):195–205. pmid:10719200
- 25. Jutkina J, Heinaru E, Vedler E, Juhanson J, Heinaru A. Occurrence of plasmids in the aromatic degrading bacterioplankton of the Baltic Sea. Genes (Basel). 2011;2(4):853–68. pmid:24710296
- 26. Viggor S, Juhanson J, Joesaar M, Mitt M, Truu J, Vedler E, et al. Dynamic changes in the structure of microbial communities in Baltic Sea coastal seawater microcosms modified by crude oil, shale oil or diesel fuel. Microbiol Res. 2013;168(7):415–27. pmid:23510642
- 27. Garibyan L, Huang T, Kim M, Wolff E, Nguyen A, Nguyen T, et al. Use of the rpoB gene to determine the specificity of base substitution mutations on the Escherichia coli chromosome. DNA Repair (Amst). 2003;2(5):593–608.
- 28. Jatsenko T, Tover A, Tegova R, Kivisaar M. Molecular characterization of Rif(r) mutations in Pseudomonas aeruginosa and Pseudomonas putida. Mutat Res. 2010;683(1–2):106–14. pmid:19887074
- 29. Bayley SA, Duggleby CJ, Worsey MJ, Williams PA, Hardy KG, Broda P. Two modes of loss of the Tol function from Pseudomonas putida mt-2. Mol Gen Genet. 1977;154(2):203–4. pmid:895716
- 30. Williams PA, Murray K. Metabolism of benzoate and the methylbenzoates by Pseudomonas putida (arvilla) mt-2: evidence for the existence of a TOL plasmid. J Bacteriol. 1974;120(1):416–23. pmid:4418209
- 31. Regenhardt D, Heuer H, Heim S, Fernandez DU, Strompl C, Moore ER, et al. Pedigree and taxonomic credentials of Pseudomonas putida strain KT2440. Environ Microbiol. 2002;4(12):912–5. pmid:12534472
- 32. Sutton MD, Smith BT, Godoy VG, Walker GC. The SOS response: recent insights into umuDC-dependent mutagenesis and DNA damage tolerance. Annu Rev Genet. 2000;34:479–97. pmid:11092836
- 33. Tang M, Bruck I, Eritja R, Turner J, Frank EG, Woodgate R, et al. Biochemical basis of SOS-induced mutagenesis in Escherichia coli: reconstitution of in vitro lesion bypass dependent on the UmuD'2C mutagenic complex and RecA protein. Proc Natl Acad Sci U S A. 1998;95(17):9755–60. pmid:9707548
- 34. Patel M, Jiang Q, Woodgate R, Cox MM, Goodman MF. A new model for SOS-induced mutagenesis: how RecA protein activates DNA polymerase V. Crit Rev Biochem Mol Biol. 2010;45(3):171–84. pmid:20441441
- 35. Tark M, Tover A, Tarassova K, Tegova R, Kivi G, Horak R, et al. A DNA polymerase V homologue encoded by TOL plasmid pWW0 confers evolutionary fitness on Pseudomonas putida under conditions of environmental stress. J Bacteriol. 2005;187(15):5203–13. pmid:16030214
- 36. Woodgate R, Sedgwick SG. Mutagenesis induced by bacterial UmuDC proteins and their plasmid homologues. Mol Microbiol. 1992;6(16):2213–8. pmid:1406263
- 37. Sundin GW, Murillo J. Functional analysis of the Pseudomonas syringae rulAB determinant in tolerance to ultraviolet B (290–320 nm) radiation and distribution of rulAB among P. syringae pathovars. Environ Microbiol. 1999;1(1):75–87. pmid:11207720
- 38. Heinaru E, Merimaa M, Viggor S, Lehiste M, Leito I, Truu J, et al. Biodegradation efficiency of functionally important populations selected for bioaugmentation in phenol- and oil-polluted area. FEMS Microbiol Ecol. 2005;51(3):363–73. pmid:16329884
- 39. Juhanson J, Truu J, Heinaru E, Heinaru A. Survival and catabolic performance of introduced Pseudomonas strains during phytoremediation and bioaugmentation field experiment. FEMS Microbiol Ecol. 2009;70(3):446–55. pmid:19732146
- 40. McHenry CS. Breaking the rules: bacteria that use several DNA polymerase IIIs. EMBO Rep. 2011;12(5):408–14. pmid:21475246
- 41. Boshoff HI, Reed MB, Barry CE, 3rd, Mizrahi V. DnaE2 polymerase contributes to in vivo survival and the emergence of drug resistance in Mycobacterium tuberculosis. Cell. 2003;113(2):183–93. pmid:12705867
- 42. Galhardo RS, Rocha RP, Marques MV, Menck CF. An SOS-regulated operon involved in damage-inducible mutagenesis in Caulobacter crescentus. Nucleic Acids Res. 2005;33(8):2603–14. pmid:15886391
- 43. Abella M, Erill I, Jara M, Mazon G, Campoy S, Barbe J. Widespread distribution of a lexA-regulated DNA damage-inducible multiple gene cassette in the Proteobacteria phylum. Mol Microbiol. 2004;54(1):212–22. pmid:15458417
- 44. Heinaru E, Vedler E, Jutkina J, Aava M, Heinaru A. Conjugal transfer and mobilization capacity of the completely sequenced naphthalene plasmid pNAH20 from multiplasmid strain Pseudomonas fluorescens PC20. FEMS Microbiol Ecol. 2009;70(3):563–74. pmid:19744238
- 45. Merimaa M, Heinaru E, Liivak M, Vedler E, Heinaru A. Grouping of phenol hydroxylase and catechol 2,3-dioxygenase genes among phenol- and p-cresol-degrading Pseudomonas species and biotypes. Arch Microbiol. 2006;186(4):287–96. pmid:16906406
- 46. Peters M, Heinaru E, Talpsep E, Wand H, Stottmeister U, Heinaru A, et al. Acquisition of a deliberately introduced phenol degradation operon, pheBA, by different indigenous Pseudomonas species. Appl Environ Microbiol. 1997;63(12):4899–906. pmid:9406411
- 47. Oliver A, Mena A. Bacterial hypermutation in cystic fibrosis, not only for antibiotic resistance. Clin Microbiol Infect. 2010;16(7):798–808. pmid:20880409
- 48. Oliver A, Baquero F, Blazquez J. The mismatch repair system (mutS, mutL and uvrD genes) in Pseudomonas aeruginosa: molecular characterization of naturally occurring mutants. Mol Microbiol. 2002;43(6):1641–50. pmid:11952911
- 49. Sikkema J, de Bont JA, Poolman B. Mechanisms of membrane toxicity of hydrocarbons. Microbiol Rev. 1995;59(2):201–22. pmid:7603409
- 50. Arjan JA, Visser M, Zeyl CW, Gerrish PJ, Blanchard JL, Lenski RE. Diminishing returns from mutation supply rate in asexual populations. Science. 1999;283(5400):404–6. pmid:9888858
- 51. Notley-McRobb L, Seeto S, Ferenci T. Enrichment and elimination of mutY mutators in Escherichia coli populations. Genetics. 2002;162(3):1055–62. pmid:12454055
- 52. Giraud A, Matic I, Tenaillon O, Clara A, Radman M, Fons M, et al. Costs and benefits of high mutation rates: adaptive evolution of bacteria in the mouse gut. Science. 2001;291(5513):2606–8. pmid:11283373
- 53. Funchain P, Yeung A, Stewart JL, Lin R, Slupska MM, Miller JH. The consequences of growth of a mutator strain of Escherichia coli as measured by loss of function among multiple gene targets and loss of fitness. Genetics. 2000;154(3):959–70. pmid:10757746
- 54. Denamur E, Lecointre G, Darlu P, Tenaillon O, Acquaviva C, Sayada C, et al. Evolutionary implications of the frequent horizontal transfer of mismatch repair genes. Cell. 2000;103(5):711–21. pmid:11114328
- 55. Brown EW, LeClerc JE, Li B, Payne WL, Cebula TA. Phylogenetic evidence for horizontal transfer of mutS alleles among naturally occurring Escherichia coli strains. J Bacteriol. 2001;183(5):1631–44. pmid:11160094
- 56. Turrientes MC, Baquero F, Levin BR, Martinez JL, Ripoll A, Gonzalez-Alba JM, et al. Normal mutation rate variants arise in a Mutator (Mut S) Escherichia coli population. PLoS One. 2013;8(9):e72963. pmid:24069167
- 57. Vidigal PG, Dittmer S, Steinmann E, Buer J, Rath PM, Steinmann J. Adaptation of Stenotrophomonas maltophilia in cystic fibrosis: molecular diversity, mutation frequency and antibiotic resistance. Int J Med Microbiol. 2014;304(5–6):613–9. pmid:24836944
- 58. Turrientes MC, Baquero MR, Sanchez MB, Valdezate S, Escudero E, Berg G, et al. Polymorphic mutation frequencies of clinical and environmental Stenotrophomonas maltophilia populations. Appl Environ Microbiol. 2010;76(6):1746–58. pmid:20097818
- 59. Baquero MR, Nilsson AI, Turrientes Mdel C, Sandvang D, Galan JC, Martinez JL, et al. Polymorphic mutation frequencies in Escherichia coli: emergence of weak mutators in clinical isolates. J Bacteriol. 2004;186(16):5538–42. pmid:15292159
- 60. Kenna DT, Doherty CJ, Foweraker J, Macaskill L, Barcus VA, Govan JR. Hypermutability in environmental Pseudomonas aeruginosa and in populations causing pulmonary infection in individuals with cystic fibrosis. Microbiology. 2007;153(Pt 6):1852–9. pmid:17526842
- 61. Lynch M. The lower bound to the evolution of mutation rates. Genome Biol Evol. 2011;3:1107–18. pmid:21821597
- 62. Lynch M, Ackerman MS, Gout JF, Long H, Sung W, Thomas WK, et al. Genetic drift, selection and the evolution of the mutation rate. Nat Rev Genet. 2016;17(11):704–14. pmid:27739533
- 63. Good BH, Desai MM. Evolution of Mutation Rates in Rapidly Adapting Asexual Populations. Genetics. 2016;204(3):1249–66. pmid:27646140
- 64. Sniegowski PD, Gerrish PJ, Johnson T, Shaver A. The evolution of mutation rates: separating causes from consequences. Bioessays. 2000;22(12):1057–66. pmid:11084621
- 65. Taverna P, Sedgwick B. Generation of an endogenous DNA-methylating agent by nitrosation in Escherichia coli. J Bacteriol. 1996;178(17):5105–11. pmid:8752326
- 66. Bjelland S, Seeberg E. Mutagenicity, toxicity and repair of DNA base damage induced by oxidation. Mutat Res. 2003;531(1–2):37–80. pmid:14637246
- 67. Foster PL. Stress-induced mutagenesis in bacteria. Crit Rev Biochem Mol Biol. 2007;42(5):373–97. pmid:17917873
- 68. Galhardo RS, Hastings PJ, Rosenberg SM. Mutation as a stress response and the regulation of evolvability. Crit Rev Biochem Mol Biol. 2007;42(5):399–435. pmid:17917874
- 69. Couce A, Blazquez J. Side effects of antibiotics on genetic variability. FEMS Microbiol Rev. 2009;33(3):531–8. pmid:19260970
- 70. Goodman MF. Error-prone repair DNA polymerases in prokaryotes and eukaryotes. Annu Rev Biochem. 2002;71:17–50. pmid:12045089
- 71. Permina EA, Mironov AA, Gelfand MS. Damage-repair error-prone polymerases of eubacteria: association with mobile genome elements. Gene. 2002;293(1–2):133–40. pmid:12137951
- 72. McBeth DL, Hauer B. Increased mutagenesis mediated by cloned plasmid CAM-OCT genes: potential for expanding substrate ranges of Pseudomonas spp. Appl Environ Microbiol. 1996;62(9):3538–40. pmid:8795249
- 73. Walker GC. Mutagenesis and inducible responses to deoxyribonucleic acid damage in Escherichia coli. Microbiol Rev. 1984;48(1):60–93. pmid:6371470
- 74. Erill I, Campoy S, Mazon G, Barbe J. Dispersal and regulation of an adaptive mutagenesis cassette in the bacteria domain. Nucleic Acids Res. 2006;34(1):66–77. pmid:16407325
- 75. Courcelle J, Khodursky A, Peter B, Brown PO, Hanawalt PC. Comparative gene expression profiles following UV exposure in wild-type and SOS-deficient Escherichia coli. Genetics. 2001;158(1):41–64. pmid:11333217
- 76. Fernandez De Henestrosa AR, Ogi T, Aoyagi S, Chafin D, Hayes JJ, Ohmori H, et al. Identification of additional genes belonging to the LexA regulon in Escherichia coli. Mol Microbiol. 2000;35(6):1560–72. pmid:10760155
- 77. Tegova R, Tover A, Tarassova K, Tark M, Kivisaar M. Involvement of error-prone DNA polymerase IV in stationary-phase mutagenesis in Pseudomonas putida. J Bacteriol. 2004;186(9):2735–44. pmid:15090515
- 78. Truglio JJ, Croteau DL, Van Houten B, Kisker C. Prokaryotic nucleotide excision repair: the UvrABC system. Chem Rev. 2006;106(2):233–52. pmid:16464004
- 79. Boyce RP, Howard-Flanders P. Release of Ultraviolet Light-Induced Thymine Dimers from DNA in E. coli K-12. Proc Natl Acad Sci U S A. 1964;51:293–300. pmid:14124327
- 80. Tsai HH, Shu HW, Yang CC, Chen CW. Translesion-synthesis DNA polymerases participate in replication of the telomeres in Streptomyces. Nucleic Acids Res. 2012;40(3):1118–30. pmid:22006845
- 81. Sidorenko J, Jatsenko T, Saumaa S, Teras R, Tark-Dame M, Hõrak R, et al. Involvement of specialized DNA polymerases Pol II, Pol IV and DnaE2 in DNA replication in the absence of Pol I in Pseudomonas putida. Mutat Res. 2011;714(1–2):63–77. pmid:21763330
- 82. Koorits L, Tegova R, Tark M, Tarassova K, Tover A, Kivisaar M. Study of involvement of ImuB and DnaE2 in stationary-phase mutagenesis in Pseudomonas putida. DNA Repair (Amst). 2007;6(6):863–8. pmid:17331811
- 83. Jatsenko T, Sidorenko J, Saumaa S, Kivisaar M. DNA Polymerases ImuC and DinB Are Involved in DNA Alkylation Damage Tolerance in Pseudomonas aeruginosa and Pseudomonas putida. PLoS One. 2017;12(1):e0170719. pmid:28118378
- 84. Robleto EA, Yasbin R, Ross C, Pedraza-Reyes M. Stationary phase mutagenesis in B. subtilis: a paradigm to study genetic diversity programs in cells under stress. Crit Rev Biochem Mol Biol. 2007;42(5):327–39. pmid:17917870
- 85. Kivisaar M. Stationary phase mutagenesis: mechanisms that accelerate adaptation of microbial populations under environmental stress. Environ Microbiol. 2003;5(10):814–27. pmid:14510835
- 86. Miller JH. A Short Course in Bacterial Genetics: A Laboratory Manual and Handbook for Escherichia coli and Related Bacteria. Cold Spring Harbour, NY: Cold Spring Harbour Laboratory Press; 1992.
- 87. Reasoner DJ, Geldreich EE. A new medium for the enumeration and subculture of bacteria from potable water. Appl Environ Microbiol. 1985;49(1):1–7. pmid:3883894
- 88. Rosche WA, Foster PL. Determining mutation rates in bacterial populations. Methods. 2000;20(1):4–17. pmid:10610800
- 89. Heinaru E, Naanuri E, Grunbach M, Joesaar M, Heinaru A. Functional redundancy in phenol and toluene degradation in Pseudomonas stutzeri strains isolated from the Baltic Sea. Gene. 2016;589(1):90–8. pmid:27185632
- 90. Hyatt D, Chen GL, Locascio PF, Land ML, Larimer FW, Hauser LJ. Prodigal: prokaryotic gene recognition and translation initiation site identification. BMC Bioinformatics. 2010;11:119. pmid:20211023
- 91. Larkin MA, Blackshields G, Brown NP, Chenna R, McGettigan PA, McWilliam H, et al. Clustal W and Clustal X version 2.0. Bioinformatics. 2007;23(21):2947–8. pmid:17846036
- 92. Page RD. TreeView: an application to display phylogenetic trees on personal computers. Comput Appl Biosci. 1996;12(4):357–8. pmid:8902363
- 93. Ruijter JM, Ramakers C, Hoogaars WM, Karlen Y, Bakker O, van den Hoff MJ, et al. Amplification efficiency: linking baseline and bias in the analysis of quantitative PCR data. Nucleic Acids Res. 2009;37(6):e45. pmid:19237396
- 94. Benjamini Y, Hochberg Y. Controlling the False Discovery Rate—a Practical and Powerful Approach to Multiple Testing. Journal of the Royal Statistical Society Series B-Methodological. 1995;57(1):289–300.
- 95. Marchler-Bauer A, Derbyshire MK, Gonzales NR, Lu S, Chitsaz F, Geer LY, et al. CDD: NCBI's conserved domain database. Nucleic Acids Res. 2015;43(Database issue):D222–6. pmid:25414356
- 96. Nohmi T, Battista JR, Dodson LA, Walker GC. RecA-mediated cleavage activates UmuD for mutagenesis: mechanistic relationship between transcriptional derepression and posttranslational activation. Proc Natl Acad Sci U S A. 1988;85(6):1816–20. pmid:3279418
- 97. Sundin GW, Kidambi SP, Ullrich M, Bender CL. Resistance to ultraviolet light in Pseudomonas syringae: sequence and functional analysis of the plasmid-encoded rulAB genes. Gene. 1996;177(1–2):77–81. pmid:8921848
- 98. Koch WH, Ennis DG, Levine AS, Woodgate R. Escherichia coli umuDC mutants: DNA sequence alterations and UmuD cleavage. Mol Gen Genet. 1992;233(3):443–8. pmid:1320188
- 99. Kato T, Shinoura Y. Isolation and characterization of mutants of Escherichia coli deficient in induction of mutations by ultraviolet light. Mol Gen Genet. 1977;156(2):121–31. pmid:340898
- 100. Steinborn G. Uvm mutants of Escherichia coli K12 deficient in UV mutagenesis. I. Isolation of uvm mutants and their phenotypical characterization in DNA repair and mutagenesis. Mol Gen Genet. 1978;165(1):87–93. pmid:362169
- 101. Woodgate R, Singh M, Kulaeva OI, Frank EG, Levine AS, Koch WH. Isolation and characterization of novel plasmid-encoded umuC mutants. J Bacteriol. 1994;176(16):5011–21. pmid:8051014
- 102. Sutton MD, Walker GC. umuDC-mediated cold sensitivity is a manifestation of functions of the UmuD(2)C complex involved in a DNA damage checkpoint control. J Bacteriol. 2001;183(4):1215–24. pmid:11157933
- 103. Shurtleff BW, Ollivierre JN, Tehrani M, Walker GC, Beuning PJ. Steric gate variants of UmuC confer UV hypersensitivity on Escherichia coli. J Bacteriol. 2009;191(15):4815–23. pmid:19482923
- 104. Hawver LA, Gillooly CA, Beuning PJ. Characterization of Escherichia coli UmuC active-site loops identifies variants that confer UV hypersensitivity. J Bacteriol. 2011;193(19):5400–11. pmid:21784925
- 105. Vaisman A, Kuban W, McDonald JP, Karata K, Yang W, Goodman MF, et al. Critical amino acids in Escherichia coli UmuC responsible for sugar discrimination and base-substitution fidelity. Nucleic Acids Res. 2012;40(13):6144–57. pmid:22422840
- 106. Boudsocq F, Ling H, Yang W, Woodgate R. Structure-based interpretation of missense mutations in Y-family DNA polymerases and their implications for polymerase function and lesion bypass. DNA Repair (Amst). 2002;1(5):343–58.
- 107. Kim DR, Pritchard AE, McHenry CS. Localization of the active site of the alpha subunit of the Escherichia coli DNA polymerase III holoenzyme. J Bacteriol. 1997;179(21):6721–8. pmid:9352922
- 108. Timinskas K, Balvociute M, Timinskas A, Venclovas C. Comprehensive analysis of DNA polymerase III alpha subunits and their homologs in bacterial genomes. Nucleic Acids Res. 2014;42(3):1393–413. pmid:24106089
- 109. Kitagawa Y, Akaboshi E, Shinagawa H, Horii T, Ogawa H, Kato T. Structural analysis of the umu operon required for inducible mutagenesis in Escherichia coli. Proc Natl Acad Sci U S A. 1985;82(13):4336–40. pmid:2989817
- 110. Wertman KF, Mount DW. Nucleotide sequence binding specificity of the LexA repressor of Escherichia coli K-12. J Bacteriol. 1985;163(1):376–84. pmid:3891738
- 111. Juurik T, Ilves H, Teras R, Ilmjarv T, Tavita K, Ukkivi K, et al. Mutation frequency and spectrum of mutations vary at different chromosomal positions of Pseudomonas putida. PLoS One. 2012;7(10):e48511. pmid:23119042
- 112. Vedler E, Heinaru E, Jutkina J, Viggor S, Koressaar T, Remm M, et al. Limnobacter spp. as newly detected phenol-degraders among Baltic Sea surface water bacteria characterised by comparative analysis of catabolic genes. Syst Appl Microbiol. 2013;36(8):525–32. pmid:24012239